Artificial Neural Networks and Deep Learning in Solid Organ Transplantation.

Journal: Transplantation
Published Date:

Abstract

Artificial neural networks are now used across many fields, from medicine to chatbots. However, they are often complicated and give "black-box" predictions. To better understand these models, we outline some of the terminology, provide example applications in solid organ transplantation, and offer recommendations for those interested in starting to use such models. Examples and definitions are sourced from PubMed searches as well as a review of several applied texts in machine and deep learning. Neural networks have been applied in transplantation since the 1990s, but more recent applications involve their use to process nontabular data, such as images and text. Although interpretations of model results should take context into account, the power of these models holds great potential for the future in transplantation.

Authors

  • Byron H Smith
    William J. von Liebig Center for Transplantation and Clinical Regeneration, Mayo Clinic, Rochester, Minnesota; and.
  • Aleksandar Denic
    Division of Nephrology and Hypertension, Mayo Clinic, Rochester, MN.

Keywords

No keywords available for this article.